This invention relates generally to database management, and particularly to optimized bulk computations in data warehouse environments.
Data warehouse systems are used to store and manage data for many purposes, such as financial, goods and services, governmental, medical, and technological industries for local, national, and global applications. For example, data warehouse systems are often used to store and manage performance data, such as wireless network performance data. Some data warehouse systems, such as performance management systems, utilize bulk computations in order to accelerate queries and reports. These computations usually pre-summarize data across several dimensions, such as time and primary key. Some computations may also involve data residing in different database tables of one or more databases and, therefore, involve joining different tables in order to derive data. Sometimes, missing or incorrect computed values can result from unexpected events, such as late or missing data or data received at different rates, which does not join using standard database techniques.